Help needed for a custom rate law

[ayush9pandey]

I would like to create a negative Hill function as my custom rate law. In the custom laws json, I currently have:

[
    {
        "name": "HillNegative", 
        "expression": "parameter / ( 1 + (reactant1 / parameter)**parameter )",
        "optional_symbols": [],
        "power_limited_species": []
    }
]

But this doesn't work and I see the following error message:

Some items in your JSON file were invalid and have been removed.
Details:
Rate law HillNegative has an invalid expression.

I have tried different versions of the above: with and without including in the optional symbols and power_limited_species list, by using a simplified expression for the hill function description, using n = 2 for the exponent hard coded..but none of these attempts have worked. I'd appreciate any help!

[longxf21]

Thanks for the issue, I've identified the bug and made a new release. Now it should work with your custom rate law

[ayush9pandey]

Thank you for pointing to the latest release (I commented on the other issue by mistake)! For some reason, pypi continues to get 0.2.7 for me. Maybe it's not cached yet. I was also going by the version in pyproject.toml on the main branch, but you may have used a different branch. Anyway, I am able to get the code with the fix.

But I still can't seem to run the model that I have. Here is my SBML file and the custom_laws.json. Please let me know what needs to be fixed on my custom_laws.json for it to verify the rate law successfully? There are so many ways in which I could write a Hill function, and I've tried a few but I keep getting "unrecognized rate law".

[longxf21]

I found that in the SBML file, the rate laws are having species that are not in the reaction.
// Reactions: r0: => A; r0_k/(1 + (C/r0_K)^r0_n); r1: => B; r1_k/(1 + (A/r1_K)^r1_n); r2: => C; r2_k/(1 + (B/r2_K)^r2_n);
The above is the reactions converted to Antimony for easier view. For r0, 'C' is in rate law. But in your custom_laws, reactant should be in rate law. Your reaction does not have a reactant. So for a reaction to be classified as HillNegative, the reaction should be:
r0: A => B; r0_k/(1 + (A/r0_K)^r0_n)
I don't know much about bio models so apologies if this is not a valid hill example.
If you want the product to be in rate law, you should change your custom_laws to be:
"expression": "parameter / ( 1 + (product1 / parameter)**parameter )",

[ayush9pandey]

Thanks for the insight! You're right that the rate is dependent on a "modifier" species that does not participate in the reaction. In SBML, this is represented as:

        <listOfModifiers>
          <modifierSpeciesReference species="C"/>
        </listOfModifiers>

for the reaction with product A:

  <reaction id="r0" name="r0" reversible="false">
        <listOfProducts>
          <speciesReference species="A" stoichiometry="1" constant="false"/>
        </listOfProducts>

Based on your explanation, the alternative that would work is updating the reactions to:

C --> A + C
A --> B + A
B --> C + B

with this update, I get a few warnings:

r0:
  Warning 1003: Flux is not increasing as reactant increases.
  Warning 1010: Irreversible reaction kinetic law contains products: C
  Warning 1031: Formatting convention not followed (compartment before parameters before species)

this I think is fair! Thank you for engaging on this issue, we can close it. Hopefully the discussion is helpful to a future user.

[luciansmith]

It seems to me that we still need to do:

• Allow the term 'modifier' in the custom rate law (unless we do already?)
• Remove warning 1003 when the total stoichiometry of the reactant in question is zero.
• Remove warning 1010 when the product in question is also a reactant.

For the first, I would like a json file like this to work:

[
    {
        "name": "HillNegative", 
        "expression": "parameter / ( 1 + (modifier1 / parameter)**parameter )",
        "optional_symbols": [],
        "power_limited_species": []
    }
]

[luciansmith]

(Should I re-open this issue, or open new ones?)

[longxf21]

Let’s open a new one, i think supporting modifier should be in future versions

…

On Mon, Jun 2, 2025 at 13:41 Lucian Smith ***@***.***> wrote: *luciansmith* left a comment (sys-bio/ratesb_python#18) <#18 (comment)> (Should I re-open this issue, or open new ones?) — Reply to this email directly, view it on GitHub <#18 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AY2M7DGGMGP6CU4RG5TDYIT3BSZGBAVCNFSM6AAAAAB5B2VUMWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDSMZSGQYTANJZGE> . You are receiving this because you modified the open/close state.Message ID: ***@***.***>

[luciansmith]

Done! #19 and #20.